@inproceedings{wu-etal-2026-bert,
title = "What Does {BERT} Know about Linguistic Features: Probing {F}innish Model for Text Difficulty",
author = "Wu, Yiheng and
Hou, Jue and
Yangarber, Roman",
editor = "Arad, Dana and
Belinkov, Yonatan and
Chen, Hanjie and
Feldhus, Nils and
Mueller, Aaron and
Mohebbi, Hosein and
Sarti, Gabriele and
Tutek, Martin",
booktitle = "Proceedings of the 9th {B}lackbox{NLP} Workshop: Analyzing and Interpreting Neural Networks for {NLP}",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.blackboxnlp-1.19/",
pages = "304--315",
ISBN = "979-8-89176-476-7",
abstract = "We observe that fine-tuned BERT outperforms feature-based models at estimation of text difficulty, but it is unclear what drives this advantage. We explore a state-of-the-art Finnish BERT difficulty estimator and analyze it against a strong feature-based model trained on 66 hand-crafted linguistic features. First, we find that BERT implicitly encodes the coarse length and complexity features, but not fine-grained morphology of Finnish. Second, BERT subsumes the information in the hand-crafted features and adds signal that they miss: the two models are evenly matched on most documents, and BERT{'}s advantage comes from rarely being very wrong, rather than from winning everywhere. Third, through input perturbation and layer-wise probing, we show that BERT{'}s predictions rely on sentence-internal structure rather than on lexical content alone. Our analysis explains why a neural difficulty estimator succeeds and what linguistic information it depends on."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wu-etal-2026-bert">
<titleInfo>
<title>What Does BERT Know about Linguistic Features: Probing Finnish Model for Text Difficulty</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yiheng</namePart>
<namePart type="family">Wu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jue</namePart>
<namePart type="family">Hou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Roman</namePart>
<namePart type="family">Yangarber</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 9th BlackboxNLP Workshop: Analyzing and Interpreting Neural Networks for NLP</title>
</titleInfo>
<name type="personal">
<namePart type="given">Dana</namePart>
<namePart type="family">Arad</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yonatan</namePart>
<namePart type="family">Belinkov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hanjie</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nils</namePart>
<namePart type="family">Feldhus</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aaron</namePart>
<namePart type="family">Mueller</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hosein</namePart>
<namePart type="family">Mohebbi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gabriele</namePart>
<namePart type="family">Sarti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Martin</namePart>
<namePart type="family">Tutek</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-476-7</identifier>
</relatedItem>
<abstract>We observe that fine-tuned BERT outperforms feature-based models at estimation of text difficulty, but it is unclear what drives this advantage. We explore a state-of-the-art Finnish BERT difficulty estimator and analyze it against a strong feature-based model trained on 66 hand-crafted linguistic features. First, we find that BERT implicitly encodes the coarse length and complexity features, but not fine-grained morphology of Finnish. Second, BERT subsumes the information in the hand-crafted features and adds signal that they miss: the two models are evenly matched on most documents, and BERT’s advantage comes from rarely being very wrong, rather than from winning everywhere. Third, through input perturbation and layer-wise probing, we show that BERT’s predictions rely on sentence-internal structure rather than on lexical content alone. Our analysis explains why a neural difficulty estimator succeeds and what linguistic information it depends on.</abstract>
<identifier type="citekey">wu-etal-2026-bert</identifier>
<location>
<url>https://aclanthology.org/2026.blackboxnlp-1.19/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>304</start>
<end>315</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T What Does BERT Know about Linguistic Features: Probing Finnish Model for Text Difficulty
%A Wu, Yiheng
%A Hou, Jue
%A Yangarber, Roman
%Y Arad, Dana
%Y Belinkov, Yonatan
%Y Chen, Hanjie
%Y Feldhus, Nils
%Y Mueller, Aaron
%Y Mohebbi, Hosein
%Y Sarti, Gabriele
%Y Tutek, Martin
%S Proceedings of the 9th BlackboxNLP Workshop: Analyzing and Interpreting Neural Networks for NLP
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-476-7
%F wu-etal-2026-bert
%X We observe that fine-tuned BERT outperforms feature-based models at estimation of text difficulty, but it is unclear what drives this advantage. We explore a state-of-the-art Finnish BERT difficulty estimator and analyze it against a strong feature-based model trained on 66 hand-crafted linguistic features. First, we find that BERT implicitly encodes the coarse length and complexity features, but not fine-grained morphology of Finnish. Second, BERT subsumes the information in the hand-crafted features and adds signal that they miss: the two models are evenly matched on most documents, and BERT’s advantage comes from rarely being very wrong, rather than from winning everywhere. Third, through input perturbation and layer-wise probing, we show that BERT’s predictions rely on sentence-internal structure rather than on lexical content alone. Our analysis explains why a neural difficulty estimator succeeds and what linguistic information it depends on.
%U https://aclanthology.org/2026.blackboxnlp-1.19/
%P 304-315
Markdown (Informal)
[What Does BERT Know about Linguistic Features: Probing Finnish Model for Text Difficulty](https://aclanthology.org/2026.blackboxnlp-1.19/) (Wu et al., BlackboxNLP 2026)
ACL